{
  "wandb_run": "run-20260630_122838-yvmohps3",
  "wandb_output_log": "/mnt/mole/SDPO/L2T/wandb/run-20260630_122838-yvmohps3/files/output.log",
  "metric_key": "val-aux/sciknoweval/reward/mean@16",
  "evals": [
    {
      "step": 10,
      "mean16": 0.4595238095238095
    },
    {
      "step": 20,
      "mean16": 0.6056547619047619
    },
    {
      "step": 30,
      "mean16": 0.6764880952380953
    },
    {
      "step": 40,
      "mean16": 0.6875
    },
    {
      "step": 50,
      "mean16": 0.7017857142857142
    },
    {
      "step": 60,
      "mean16": 0.7232142857142857
    },
    {
      "step": 70,
      "mean16": 0.7160714285714286
    },
    {
      "step": 80,
      "mean16": 0.7220238095238095
    },
    {
      "step": 90,
      "mean16": 0.731845238095238
    },
    {
      "step": 100,
      "mean16": 0.74375
    }
  ],
  "best_step": 100,
  "best_mean16": 0.74375,
  "final_step": 100,
  "final_mean16": 0.74375
}